Phonology, Semantics, And The Comprehension-expression Gap In Emerging Lexicons

Abstract

Purpose: Children come to understand many words by the end of their first year of life, and yet, generally by 12 months only a few words are said. In this study, we investigated which linguistic factors contribute to this comprehension-expression gap the most. Specifically, we asked: are phonological neighborhood density (PND), semantic neighborhood density (SND), and word frequency (WF) significant predictors of the probability that words known (understood) by children would appear in their spoken lexicons? Method: Monosyllabic words in the active (understood and said) and passive (understood, not said) lexicons of 201 toddlers were extracted from the Dutch Communicative Development Inventory (NCDI) parent-completed forms. A Generalised Linear Mixed Effects Model (GLMEM) was applied to the data. Results: PND and WF were independently and significantly associated with whether or not a known word would be in children’s spoken lexicons but SND was not. There were individual differences in the impact of WF on the probability that known words would be said. Conclusion: The novel findings reported here have two major implications. First, they indicate that the comprehension-expression gap exists partly because the phonological distributional properties of words determine how readily words can be phonologically encoded for word production. Second, there are likely subtle and complex individual differences in how and when the statistical properties of the ambient language impact on children’s emerging lexicons that might best be explored via longitudinal sampling of word knowledge and use.